refactor: rename regexp.useJVM boolean to regexp.engine enum (rust|java)

Author: andygrove

common/src/main/scala/org/apache/comet/CometConf.scala

@@ -383,17 +383,19 @@ object CometConf extends ShimCometConf {
       .booleanConf
       .createWithDefault(false)
 
-  val COMET_REGEXP_USE_JVM: ConfigEntry[Boolean] =
-    conf("spark.comet.exec.regexp.useJVM")
+  val COMET_REGEXP_ENGINE: ConfigEntry[String] =
+    conf("spark.comet.exec.regexp.engine")
       .category(CATEGORY_EXEC)
       .doc(
-        "Experimental. When true, route supported regular-expression expressions through a " +
-          "JVM-side UDF (java.util.regex.Pattern) for Spark-compatible semantics, at the cost " +
-          "of JNI roundtrips per batch. When false, falls back to whichever native or Spark " +
-          "path the expression normally selects. Only RLike is routed today; additional " +
+        "Experimental. Selects the engine used to evaluate supported regular-expression " +
+          "expressions. `rust` uses the native DataFusion regexp engine. `java` routes through " +
+          "a JVM-side UDF (java.util.regex.Pattern) for Spark-compatible semantics, at the " +
+          "cost of JNI roundtrips per batch. Only RLike is routed today; additional " +
           "expressions may opt in over time.")
-      .booleanConf
-      .createWithDefault(false)
+      .stringConf
+      .transform(_.toLowerCase(Locale.ROOT))
+      .checkValues(Set("rust", "java"))
+      .createWithDefault("rust")
 
   val COMET_EXEC_SHUFFLE_WITH_HASH_PARTITIONING_ENABLED: ConfigEntry[Boolean] =
     conf("spark.comet.native.shuffle.partitioning.hash.enabled")

spark/src/main/scala/org/apache/comet/serde/strings.scala

@@ -267,15 +267,15 @@ object CometRLike extends CometExpressionSerde[RLike] {
     "Uses Rust regexp engine, which has different behavior to Java regexp engine")
 
   override def getSupportLevel(expr: RLike): SupportLevel = {
-    if (CometConf.COMET_REGEXP_USE_JVM.get()) {
+    if (CometConf.COMET_REGEXP_ENGINE.get() == "java") {
       Compatible(None)
     } else {
       super.getSupportLevel(expr)
     }
   }
 
   override def convert(expr: RLike, inputs: Seq[Attribute], binding: Boolean): Option[Expr] = {
-    if (CometConf.COMET_REGEXP_USE_JVM.get()) {
+    if (CometConf.COMET_REGEXP_ENGINE.get() == "java") {
       convertViaJvmUdf(expr, inputs, binding)
     } else {
       convertViaNativeRegex(expr, inputs, binding)

spark/src/test/scala/org/apache/comet/CometRegExpJvmSuite.scala

@@ -27,7 +27,7 @@ import org.apache.spark.sql.execution.adaptive.AdaptiveSparkPlanHelper
 class CometRegExpJvmSuite extends CometTestBase with AdaptiveSparkPlanHelper {
 
   override protected def sparkConf: SparkConf =
-    super.sparkConf.set("spark.comet.exec.regexp.useJVM", "true")
+    super.sparkConf.set("spark.comet.exec.regexp.engine", "java")
 
   test("rlike: Java regex semantics, with flag on") {
     withTable("t") {

spark/src/test/scala/org/apache/spark/sql/benchmark/CometRegExpBenchmark.scala

@@ -110,7 +110,7 @@ object CometRegExpBenchmark extends CometBenchmarkBase {
     }
 
     benchmark.addCase("Comet (Exec, JVM regex)") { _ =>
-      val configs = baseExec ++ Map(CometConf.COMET_REGEXP_USE_JVM.key -> "true")
+      val configs = baseExec ++ Map(CometConf.COMET_REGEXP_ENGINE.key -> "java")
       withSQLConf(configs.toSeq: _*) {
         spark.sql(query).noop()
       }